Atrial fibrillation (AF) is the most common cardiac arrhythmia in the United States, and imposes a high risk of embolic stroke. Alcohol, the most commonly consumed drug in the United States, has been considered to be a precipitant of the arrhythmia. While the relationship between alcohol and AF has been accepted based on observational reports, prior studies have all relied on patient self-report of alcohol consumption and insensitive markers of AF, such as hospitalization, to assess the relationship between the two. Therefore, to confirm and understand the degree of association between alcohol consumption and AF, more reliable methods are needed. Furthermore, prior published studies demonstrating an increased risk of AF with alcohol consumption have not been powered to measure the potential effect modification of patient characteristics or comorbid diseases. It is unknown if particular patient characteristics are protective in the face of alcohol, or if certain comorbidities place patients at increased risk for developing AF. The overall hypothesis of our project is that there is an association between alcohol and AF in the acute phase of even moderate consumption in patients with paroxysmal AF, and that the degree of the relationship between alcohol and AF depends on certain patient characteristics. This hypothesis will be tested by the following aims: Aim 1 will determine the relationship between acute alcohol consumption and incidence of paroxysms of AF, as measured by gold standard continuous monitoring devices for blood alcohol levels and ambulatory arrhythmia detection. This will provide insight into the association at a level of precision in both alcohol consumption and detection of arrhythmia never before published. Aim 2 will determine the currently unknown patient characteristics that modify the relationship between alcohol and AF on a population level.